_ Git - youtube-dl/blob - youtube_dl/extractor/arte.py

   1 # encoding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..compat import (
   8     compat_parse_qs,
   9     compat_urllib_parse_urlparse,
  10 )
  11 from ..utils import (
  12     find_xpath_attr,
  13     unified_strdate,
  14     get_element_by_attribute,
  15     int_or_none,
  16     NO_DEFAULT,
  17     qualities,
  18 )
  19
  20 # There are different sources of video in arte.tv, the extraction process
  21 # is different for each one. The videos usually expire in 7 days, so we can't
  22 # add tests.
  23
  24
  25 class ArteTvIE(InfoExtractor):
  26     _VALID_URL = r'https?://videos\.arte\.tv/(?P<lang>fr|de|en|es)/.*-(?P<id>.*?)\.html'
  27     IE_NAME = 'arte.tv'
  28
  29     def _real_extract(self, url):
  30         mobj = re.match(self._VALID_URL, url)
  31         lang = mobj.group('lang')
  32         video_id = mobj.group('id')
  33
  34         ref_xml_url = url.replace('/videos/', '/do_delegate/videos/')
  35         ref_xml_url = ref_xml_url.replace('.html', ',view,asPlayerXml.xml')
  36         ref_xml_doc = self._download_xml(
  37             ref_xml_url, video_id, note='Downloading metadata')
  38         config_node = find_xpath_attr(ref_xml_doc, './/video', 'lang', lang)
  39         config_xml_url = config_node.attrib['ref']
  40         config = self._download_xml(
  41             config_xml_url, video_id, note='Downloading configuration')
  42
  43         formats = [{
  44             'format_id': q.attrib['quality'],
  45             # The playpath starts at 'mp4:', if we don't manually
  46             # split the url, rtmpdump will incorrectly parse them
  47             'url': q.text.split('mp4:', 1)[0],
  48             'play_path': 'mp4:' + q.text.split('mp4:', 1)[1],
  49             'ext': 'flv',
  50             'quality': 2 if q.attrib['quality'] == 'hd' else 1,
  51         } for q in config.findall('./urls/url')]
  52         self._sort_formats(formats)
  53
  54         title = config.find('.//name').text
  55         thumbnail = config.find('.//firstThumbnailUrl').text
  56         return {
  57             'id': video_id,
  58             'title': title,
  59             'thumbnail': thumbnail,
  60             'formats': formats,
  61         }
  62
  63
  64 class ArteTVBaseIE(InfoExtractor):
  65     @classmethod
  66     def _extract_url_info(cls, url):
  67         mobj = re.match(cls._VALID_URL, url)
  68         lang = mobj.group('lang')
  69         query = compat_parse_qs(compat_urllib_parse_urlparse(url).query)
  70         if 'vid' in query:
  71             video_id = query['vid'][0]
  72         else:
  73             # This is not a real id, it can be for example AJT for the news
  74             # http://www.arte.tv/guide/fr/emissions/AJT/arte-journal
  75             video_id = mobj.group('id')
  76         return video_id, lang
  77
  78     def _extract_from_json_url(self, json_url, video_id, lang, title=None):
  79         info = self._download_json(json_url, video_id)
  80         player_info = info['videoJsonPlayer']
  81
  82         upload_date_str = player_info.get('shootingDate')
  83         if not upload_date_str:
  84             upload_date_str = (player_info.get('VRA') or player_info.get('VDA') or '').split(' ')[0]
  85
  86         title = (player_info.get('VTI') or title or player_info['VID']).strip()
  87         subtitle = player_info.get('VSU', '').strip()
  88         if subtitle:
  89             title += ' - %s' % subtitle
  90
  91         info_dict = {
  92             'id': player_info['VID'],
  93             'title': title,
  94             'description': player_info.get('VDE'),
  95             'upload_date': unified_strdate(upload_date_str),
  96             'thumbnail': player_info.get('programImage') or player_info.get('VTU', {}).get('IUR'),
  97         }
  98         qfunc = qualities(['HQ', 'MQ', 'EQ', 'SQ'])
  99
 100         LANGS = {
 101             'fr': 'F',
 102             'de': 'A',
 103             'en': 'E[ANG]',
 104             'es': 'E[ESP]',
 105         }
 106
 107         langcode = LANGS.get(lang, lang)
 108
 109         formats = []
 110         for format_id, format_dict in player_info['VSR'].items():
 111             f = dict(format_dict)
 112             versionCode = f.get('versionCode')
 113             l = re.escape(langcode)
 114
 115             # Language preference from most to least priority
 116             # Reference: section 5.6.3 of
 117             # http://www.arte.tv/sites/en/corporate/files/complete-technical-guidelines-arte-geie-v1-05.pdf
 118             PREFERENCES = (
 119                 # original version in requested language, without subtitles
 120                 r'VO{0}$'.format(l),
 121                 # original version in requested language, with partial subtitles in requested language
 122                 r'VO{0}-ST{0}$'.format(l),
 123                 # original version in requested language, with subtitles for the deaf and hard-of-hearing in requested language
 124                 r'VO{0}-STM{0}$'.format(l),
 125                 # non-original (dubbed) version in requested language, without subtitles
 126                 r'V{0}$'.format(l),
 127                 # non-original (dubbed) version in requested language, with subtitles partial subtitles in requested language
 128                 r'V{0}-ST{0}$'.format(l),
 129                 # non-original (dubbed) version in requested language, with subtitles for the deaf and hard-of-hearing in requested language
 130                 r'V{0}-STM{0}$'.format(l),
 131                 # original version in requested language, with partial subtitles in different language
 132                 r'VO{0}-ST(?!{0}).+?$'.format(l),
 133                 # original version in requested language, with subtitles for the deaf and hard-of-hearing in different language
 134                 r'VO{0}-STM(?!{0}).+?$'.format(l),
 135                 # original version in different language, with partial subtitles in requested language
 136                 r'VO(?:(?!{0}).+?)?-ST{0}$'.format(l),
 137                 # original version in different language, with subtitles for the deaf and hard-of-hearing in requested language
 138                 r'VO(?:(?!{0}).+?)?-STM{0}$'.format(l),
 139                 # original version in different language, without subtitles
 140                 r'VO(?:(?!{0}))?$'.format(l),
 141                 # original version in different language, with partial subtitles in different language
 142                 r'VO(?:(?!{0}).+?)?-ST(?!{0}).+?$'.format(l),
 143                 # original version in different language, with subtitles for the deaf and hard-of-hearing in different language
 144                 r'VO(?:(?!{0}).+?)?-STM(?!{0}).+?$'.format(l),
 145             )
 146
 147             for pref, p in enumerate(PREFERENCES):
 148                 if re.match(p, versionCode):
 149                     lang_pref = len(PREFERENCES) - pref
 150                     break
 151             else:
 152                 lang_pref = -1
 153
 154             format = {
 155                 'format_id': format_id,
 156                 'preference': -10 if f.get('videoFormat') == 'M3U8' else None,
 157                 'language_preference': lang_pref,
 158                 'format_note': '%s, %s' % (f.get('versionCode'), f.get('versionLibelle')),
 159                 'width': int_or_none(f.get('width')),
 160                 'height': int_or_none(f.get('height')),
 161                 'tbr': int_or_none(f.get('bitrate')),
 162                 'quality': qfunc(f.get('quality')),
 163             }
 164
 165             if f.get('mediaType') == 'rtmp':
 166                 format['url'] = f['streamer']
 167                 format['play_path'] = 'mp4:' + f['url']
 168                 format['ext'] = 'flv'
 169             else:
 170                 format['url'] = f['url']
 171
 172             formats.append(format)
 173
 174         self._check_formats(formats, video_id)
 175         self._sort_formats(formats)
 176
 177         info_dict['formats'] = formats
 178         return info_dict
 179
 180
 181 class ArteTVPlus7IE(ArteTVBaseIE):
 182     IE_NAME = 'arte.tv:+7'
 183     _VALID_URL = r'https?://(?:www\.)?arte\.tv/guide/(?P<lang>fr|de|en|es)/(?:(?:sendungen|emissions|embed)/)?(?P<id>[^/]+)/(?P<name>[^/?#&]+)'
 184
 185     _TESTS = [{
 186         'url': 'http://www.arte.tv/guide/de/sendungen/XEN/xenius/?vid=055918-015_PLUS7-D',
 187         'only_matching': True,
 188     }]
 189
 190     @classmethod
 191     def suitable(cls, url):
 192         return False if ArteTVPlaylistIE.suitable(url) else super(ArteTVPlus7IE, cls).suitable(url)
 193
 194     def _real_extract(self, url):
 195         video_id, lang = self._extract_url_info(url)
 196         webpage = self._download_webpage(url, video_id)
 197         return self._extract_from_webpage(webpage, video_id, lang)
 198
 199     def _extract_from_webpage(self, webpage, video_id, lang):
 200         patterns_templates = (r'arte_vp_url=["\'](.*?%s.*?)["\']', r'data-url=["\']([^"]+%s[^"]+)["\']')
 201         ids = (video_id, '')
 202         # some pages contain multiple videos (like
 203         # http://www.arte.tv/guide/de/sendungen/XEN/xenius/?vid=055918-015_PLUS7-D),
 204         # so we first try to look for json URLs that contain the video id from
 205         # the 'vid' parameter.
 206         patterns = [t % re.escape(_id) for _id in ids for t in patterns_templates]
 207         json_url = self._html_search_regex(
 208             patterns, webpage, 'json vp url', default=None)
 209         if not json_url:
 210             def find_iframe_url(webpage, default=NO_DEFAULT):
 211                 return self._html_search_regex(
 212                     r'<iframe[^>]+src=(["\'])(?P<url>.+\bjson_url=.+?)\1',
 213                     webpage, 'iframe url', group='url', default=default)
 214
 215             iframe_url = find_iframe_url(webpage, None)
 216             if not iframe_url:
 217                 embed_url = self._html_search_regex(
 218                     r'arte_vp_url_oembed=\'([^\']+?)\'', webpage, 'embed url', default=None)
 219                 if embed_url:
 220                     player = self._download_json(
 221                         embed_url, video_id, 'Downloading player page')
 222                     iframe_url = find_iframe_url(player['html'])
 223             # en and es URLs produce react-based pages with different layout (e.g.
 224             # http://www.arte.tv/guide/en/053330-002-A/carnival-italy?zone=world)
 225             if not iframe_url:
 226                 program = self._search_regex(
 227                     r'program\s*:\s*({.+?["\']embed_html["\'].+?}),?\s*\n',
 228                     webpage, 'program', default=None)
 229                 if program:
 230                     embed_html = self._parse_json(program, video_id)
 231                     if embed_html:
 232                         iframe_url = find_iframe_url(embed_html['embed_html'])
 233             if iframe_url:
 234                 json_url = compat_parse_qs(
 235                     compat_urllib_parse_urlparse(iframe_url).query)['json_url'][0]
 236         if json_url:
 237             title = self._search_regex(
 238                 r'<h3[^>]+title=(["\'])(?P<title>.+?)\1',
 239                 webpage, 'title', default=None, group='title')
 240             return self._extract_from_json_url(json_url, video_id, lang, title=title)
 241         # Different kind of embed URL (e.g.
 242         # http://www.arte.tv/magazine/trepalium/fr/episode-0406-replay-trepalium)
 243         entries = [
 244             self.url_result(url)
 245             for _, url in re.findall(r'<iframe[^>]+src=(["\'])(?P<url>.+?)\1', webpage)]
 246         return self.playlist_result(entries)
 247
 248
 249 # It also uses the arte_vp_url url from the webpage to extract the information
 250 class ArteTVCreativeIE(ArteTVPlus7IE):
 251     IE_NAME = 'arte.tv:creative'
 252     _VALID_URL = r'https?://creative\.arte\.tv/(?P<lang>fr|de|en|es)/(?:[^/]+/)*(?P<id>[^/?#&]+)'
 253
 254     _TESTS = [{
 255         'url': 'http://creative.arte.tv/fr/episode/osmosis-episode-1',
 256         'info_dict': {
 257             'id': '057405-001-A',
 258             'ext': 'mp4',
 259             'title': 'OSMOSIS - N\'AYEZ PLUS PEUR D\'AIMER (1)',
 260             'upload_date': '20150716',
 261         },
 262     }, {
 263         'url': 'http://creative.arte.tv/fr/Monty-Python-Reunion',
 264         'playlist_count': 11,
 265         'add_ie': ['Youtube'],
 266     }, {
 267         'url': 'http://creative.arte.tv/de/episode/agentur-amateur-4-der-erste-kunde',
 268         'only_matching': True,
 269     }]
 270
 271
 272 class ArteTVInfoIE(ArteTVPlus7IE):
 273     IE_NAME = 'arte.tv:info'
 274     _VALID_URL = r'https?://info\.arte\.tv/(?P<lang>fr|de|en|es)/(?:[^/]+/)*(?P<id>[^/?#&]+)'
 275
 276     _TESTS = [{
 277         'url': 'http://info.arte.tv/fr/service-civique-un-cache-misere',
 278         'info_dict': {
 279             'id': '067528-000-A',
 280             'ext': 'mp4',
 281             'title': 'Service civique, un cache misère ?',
 282             'upload_date': '20160403',
 283         },
 284     }]
 285
 286
 287 class ArteTVFutureIE(ArteTVPlus7IE):
 288     IE_NAME = 'arte.tv:future'
 289     _VALID_URL = r'https?://future\.arte\.tv/(?P<lang>fr|de|en|es)/(?P<id>[^/?#&]+)'
 290
 291     _TESTS = [{
 292         'url': 'http://future.arte.tv/fr/info-sciences/les-ecrevisses-aussi-sont-anxieuses',
 293         'info_dict': {
 294             'id': '050940-028-A',
 295             'ext': 'mp4',
 296             'title': 'Les écrevisses aussi peuvent être anxieuses',
 297             'upload_date': '20140902',
 298         },
 299     }, {
 300         'url': 'http://future.arte.tv/fr/la-science-est-elle-responsable',
 301         'only_matching': True,
 302     }]
 303
 304
 305 class ArteTVDDCIE(ArteTVPlus7IE):
 306     IE_NAME = 'arte.tv:ddc'
 307     _VALID_URL = r'https?://ddc\.arte\.tv/(?P<lang>emission|folge)/(?P<id>[^/?#&]+)'
 308
 309     _TESTS = []
 310
 311     def _real_extract(self, url):
 312         video_id, lang = self._extract_url_info(url)
 313         if lang == 'folge':
 314             lang = 'de'
 315         elif lang == 'emission':
 316             lang = 'fr'
 317         webpage = self._download_webpage(url, video_id)
 318         scriptElement = get_element_by_attribute('class', 'visu_video_block', webpage)
 319         script_url = self._html_search_regex(r'src="(.*?)"', scriptElement, 'script url')
 320         javascriptPlayerGenerator = self._download_webpage(script_url, video_id, 'Download javascript player generator')
 321         json_url = self._search_regex(r"json_url=(.*)&rendering_place.*", javascriptPlayerGenerator, 'json url')
 322         return self._extract_from_json_url(json_url, video_id, lang)
 323
 324
 325 class ArteTVConcertIE(ArteTVPlus7IE):
 326     IE_NAME = 'arte.tv:concert'
 327     _VALID_URL = r'https?://concert\.arte\.tv/(?P<lang>fr|de|en|es)/(?P<id>[^/?#&]+)'
 328
 329     _TESTS = [{
 330         'url': 'http://concert.arte.tv/de/notwist-im-pariser-konzertclub-divan-du-monde',
 331         'md5': '9ea035b7bd69696b67aa2ccaaa218161',
 332         'info_dict': {
 333             'id': '186',
 334             'ext': 'mp4',
 335             'title': 'The Notwist im Pariser Konzertclub "Divan du Monde"',
 336             'upload_date': '20140128',
 337             'description': 'md5:486eb08f991552ade77439fe6d82c305',
 338         },
 339     }]
 340
 341
 342 class ArteTVCinemaIE(ArteTVPlus7IE):
 343     IE_NAME = 'arte.tv:cinema'
 344     _VALID_URL = r'https?://cinema\.arte\.tv/(?P<lang>fr|de|en|es)/(?P<id>.+)'
 345
 346     _TESTS = [{
 347         'url': 'http://cinema.arte.tv/fr/article/les-ailes-du-desir-de-julia-reck',
 348         'md5': 'a5b9dd5575a11d93daf0e3f404f45438',
 349         'info_dict': {
 350             'id': '062494-000-A',
 351             'ext': 'mp4',
 352             'title': 'Film lauréat du concours web - "Les ailes du désir" de Julia Reck',
 353             'upload_date': '20150807',
 354         },
 355     }]
 356
 357
 358 class ArteTVMagazineIE(ArteTVPlus7IE):
 359     IE_NAME = 'arte.tv:magazine'
 360     _VALID_URL = r'https?://(?:www\.)?arte\.tv/magazine/[^/]+/(?P<lang>fr|de|en|es)/(?P<id>[^/?#&]+)'
 361
 362     _TESTS = [{
 363         # Embedded via <iframe src="http://www.arte.tv/arte_vp/index.php?json_url=..."
 364         'url': 'http://www.arte.tv/magazine/trepalium/fr/entretien-avec-le-realisateur-vincent-lannoo-trepalium',
 365         'md5': '2a9369bcccf847d1c741e51416299f25',
 366         'info_dict': {
 367             'id': '065965-000-A',
 368             'ext': 'mp4',
 369             'title': 'Trepalium - Extrait Ep.01',
 370             'upload_date': '20160121',
 371         },
 372     }, {
 373         # Embedded via <iframe src="http://www.arte.tv/guide/fr/embed/054813-004-A/medium"
 374         'url': 'http://www.arte.tv/magazine/trepalium/fr/episode-0406-replay-trepalium',
 375         'md5': 'fedc64fc7a946110fe311634e79782ca',
 376         'info_dict': {
 377             'id': '054813-004_PLUS7-F',
 378             'ext': 'mp4',
 379             'title': 'Trepalium (4/6)',
 380             'description': 'md5:10057003c34d54e95350be4f9b05cb40',
 381             'upload_date': '20160218',
 382         },
 383     }, {
 384         'url': 'http://www.arte.tv/magazine/metropolis/de/frank-woeste-german-paris-metropolis',
 385         'only_matching': True,
 386     }]
 387
 388
 389 class ArteTVEmbedIE(ArteTVPlus7IE):
 390     IE_NAME = 'arte.tv:embed'
 391     _VALID_URL = r'''(?x)
 392         http://www\.arte\.tv
 393         /(?:playerv2/embed|arte_vp/index)\.php\?json_url=
 394         (?P<json_url>
 395             http://arte\.tv/papi/tvguide/videos/stream/player/
 396             (?P<lang>[^/]+)/(?P<id>[^/]+)[^&]*
 397         )
 398     '''
 399
 400     _TESTS = []
 401
 402     def _real_extract(self, url):
 403         mobj = re.match(self._VALID_URL, url)
 404         video_id = mobj.group('id')
 405         lang = mobj.group('lang')
 406         json_url = mobj.group('json_url')
 407         return self._extract_from_json_url(json_url, video_id, lang)
 408
 409
 410 class ArteTVPlaylistIE(ArteTVBaseIE):
 411     IE_NAME = 'arte.tv:playlist'
 412     _VALID_URL = r'https?://(?:www\.)?arte\.tv/guide/(?P<lang>fr|de|en|es)/[^#]*#collection/(?P<id>PL-\d+)'
 413
 414     _TESTS = [{
 415         'url': 'http://www.arte.tv/guide/de/plus7/?country=DE#collection/PL-013263/ARTETV',
 416         'info_dict': {
 417             'id': 'PL-013263',
 418             'title': 'Areva & Uramin',
 419         },
 420         'playlist_mincount': 6,
 421     }, {
 422         'url': 'http://www.arte.tv/guide/de/playlists?country=DE#collection/PL-013190/ARTETV',
 423         'only_matching': True,
 424     }]
 425
 426     def _real_extract(self, url):
 427         playlist_id, lang = self._extract_url_info(url)
 428         collection = self._download_json(
 429             'https://api.arte.tv/api/player/v1/collectionData/%s/%s?source=videos'
 430             % (lang, playlist_id), playlist_id)
 431         title = collection.get('title')
 432         description = collection.get('shortDescription') or collection.get('teaserText')
 433         entries = [
 434             self._extract_from_json_url(
 435                 video['jsonUrl'], video.get('programId') or playlist_id, lang)
 436             for video in collection['videos'] if video.get('jsonUrl')]
 437         return self.playlist_result(entries, playlist_id, title, description)